Application Ser. No. 09/084,042 of Matthew P. Mitchell for Concentric Foil Structure for Regenerators.
1. Backgroundxe2x80x94Field of Invention
The invention relates to liners for the walls of the pulse tube portions of pulse tube refrigerators.
2. Backgroundxe2x80x94Description of Prior Art
Pulse tube refrigerators are regenerative gas cycle refrigerators typically used as cryocoolers, providing cooling at temperatures below about 120 Kelvin. Pulse tube refrigerators are characterized by a tube, called the xe2x80x9cpulse tubexe2x80x9d in which a compressible fluid, typically helium, is cyclically shuttled back and forth while the pressure of the fluid, and thus its temperature, is cyclically changing. One end of the pulse tube becomes warm as warm, compressed fluid repeatedly moves toward the warm heat exchanger, where heat is rejected. The other end of the pulse tube becomes cold as fluid at lower pressure repeatedly moves toward the cold heat exchanger, where heat is lifted from the cooling load. In operation, fluid in the pulse tube acquires a temperature gradient from one end of the pulse tube to the other end. The wall of the pulse tube likewise acquires a temperature gradient from its warm end to its cold end. However, due to movement of the fluid, the temperature at any point on the wall of the pulse tube is seldom the same as the temperature of the fluid in contact with it.
With in-line and U-tube configurations, the pulse tube must be strong enough to contain the internal pressure of the working fluid with a margin of safety. It must also be strong enough to handle the mechanical stresses that it will experience during assembly and operation. That ordinarily implies a minimum metal wall thickness of the order of 0.3 mm for refrigerators with a few watts of capacity and thicker walls for larger machines. Because metals have high diffusivity and substantial volumetric heat capacity, their thermal inertia is high. Thus pulse tube walls have much more thermal mass than the working fluid in the refrigerator, and their local temperatures change little over the course of a cycle. Heat transfer between the working fluid, in which temperature is constantly changing, and pulse tube walls, which remain essentially isothermal, seriously damps temperature swings in the fluid, especially in small-diameter pulse tubes, in which much of the fluid lies within a penetration depth from the wall of the pulse tube, and in low frequency refrigerators, in which heat transfer occurs over a relatively long time interval, thereby increasing penetration depths.
Heat transfer between fluid and pulse tube wall also tends to generate a xe2x80x9cstreamingxe2x80x9d effect in the fluid in the pulse tube. Streaming causes fluid adjacent to the wall of the pulse tube to move toward its warm end; a balancing flow at the axis of the pulse tube moves from the warm end toward the cold end. Torroidal convection generated by streaming flows constitutes another loss mechanism that decreases cooling power and reduces efficiency of the refrigerator.
The adverse effects of temperature-swing damping and streaming have been recognized by others. A solution to the streaming problem been proposed. Olson and Swift have counteracted streaming with a carefully-calculated taper in pulse tube walls. (U.S. Pat. No. 5,953,920). That, however, does not prevent the adverse effects of heat transfer in damping temperature swings in the fluid.
Marquardt and Radebaugh have suggested the use of plastic liners in a pulse tube as a means of changing the volume of the pulse tube, and to reduce conduction losses. They also mention the possibility of tapering the liners to reduce streaming. (xe2x80x9cPulse Tube Oxygen Liquefierxe2x80x9d, Advances in Cryogenic Engineering, Vol. 45A, p. 457 at p. 460 (Kluwer Academic/Plenum Publishers 1999)). While not expressly noted by Marquardt and Radebaugh, the relatively poor heat transfer in plastic would permit its surface temperature to fluctuate somewhat more than would the wall of a metal pulse tube of equal thickness. However, the volumetric heat capacity of suitable plastic materials is substantial, and a plastic liner would need to be relatively thick to provide the structural strength required survive handling and to maintain its integrity in place. That would require substantial thermal mass in a plastic liner, providing no adequate solution to the temperature-swing damping problem. Moreover, the coefficients of expansion for plastic materials are substantially larger than for metals; the cold end of a plastic liner would contract more than a steel pulse tube in which it was installed, opening up a gap that would create undesirable xe2x80x9cappendix gapxe2x80x9d losses well understood in the Stirling Cycle engine art. No successful application of plastic pulse tube liners has been reported.
A thin liner fabricated from a strong material with relatively low heat capacity, preferably of metal, is installed in the pulse tube in close proximity to the pulse tube wall. Because the liner is in intimate contact with the fluid in the pulse tube, and because the fluid in the pulse tube is almost always in motion, heat transfer between the fluid and the liner is relatively good. Because the liner is in only intermittent contact with the wall of the pulse tube, and because the thin layer of fluid trapped between the liner and the wall of the pulse tube is stagnant, heat transfer between the liner and the wall of the pulse tube is relatively poor. Because the liner itself is thin, its heat capacity is low. Thus, the effect of heat transfers between the fluid and the liner is to substantially alter the temperature of the liner as well as the temperature of the fluid, raising the liner temperature as the fluid is cooled, and vice versa. The result is that the temperature difference between the fluid and the liner at any instant is less than it would otherwise be, and thus less heat is transferred back and forth between the liner and the fluid, resulting in a smaller change in the temperature of the fluid. That reduces thermodynamic losses due to damping of temperature swings in the fluid and reduces the tendency toward streaming that would otherwise occur in untapered pulse tubes.
Several objects and advantages of this invention are:
(1) To reduce thermodynamic losses resulting from the damping effect of heat transfer between the pulse tube of a pulse tube refrigerator and the fluid in that pulse tube.
(2) To reduce thermodynamic losses resulting from streaming effects induced by heat transfer between the pulse tube of a pulse tube refrigerator and the fluid in that pulse tube.
(3) To provide simple, inexpensive means for reducing thermodynamic losses resulting from the damping effect of heat transfer between the pulse tube of a pulse tube refrigerator and the fluid in that pulse tube.
(4) To provide simple, inexpensive means for reducing thermodynamic losses resulting from streaming effects induced by heat transfer between the pulse tube of a pulse tube refrigerator and the fluid in that pulse tube.
Further objects and advantages will become apparent from a consideration of the following description and drawings.